.map file data - meaning of brush face coordinates

[Clueless Newbie]

I made a .map file with one caulked cubic brush and saved it, and then opened it in a text editor. This is what the file looks like:

Code: Select all

// entity 0
{
"classname" "worldspawn"
// brush 0
{
( 64 64 0 ) ( 0 64 0 ) ( 0 0 0 ) common/caulk 0 0 0 0.500000 0.500000 0 4 0
( 0 0 64 ) ( 0 64 64 ) ( 64 64 64 ) common/caulk 0 0 0 0.500000 0.500000 0 4 0
( 0 0 64 ) ( 64 0 64 ) ( 64 0 0 ) common/caulk 0 0 0 0.500000 0.500000 0 4 0
( 64 0 64 ) ( 64 64 64 ) ( 64 64 0 ) common/caulk 0 0 0 0.500000 0.500000 0 4 0
( 64 64 64 ) ( 0 64 64 ) ( 0 64 0 ) common/caulk 0 0 0 0.500000 0.500000 0 4 0
( 0 64 64 ) ( 0 0 64 ) ( 0 0 0 ) common/caulk 0 0 0 0.500000 0.500000 0 4 0
}
}



I assume there is one data set for each side. But what are the coordinates? Three of the four corners on a given side? If so, which of the corners? And what are the texture coordinates? The same three corners (textures are 2d, so only two nrs pr coord) plus two more values? Scale and rotation, maybe?

[divVerent]

No.

A brush is described not by vertices, but by planes.

( Ax Ay Az ) (Bx By Bz) (Cx Cy Cz) Texture ShiftS ShiftT RotationAngle ScaleS ScaleT Flags Ignored Ignored

That describes the plane on which the triangle ABC lies, which gets a texture... and the Shift, Rotation and Scale parameters are exactly like in the GtkRadiant surface properties. The Flags field only uses one bit to describe whether this is a detail brush or not.

The points ABC are not necessarily actual brush vertices. They CAN be, but often they aren't. To get the actual vertices, you need to intersect the planes with others.

[Clueless Newbie]

No.

A brush is described not by vertices, but by planes.

( Ax Ay Az ) (Bx By Bz) (Cx Cy Cz) Texture ShiftS ShiftT RotationAngle ScaleS ScaleT Flags Ignored Ignored

That describes the plane on which the triangle ABC lies, which gets a texture... and the Shift, Rotation and Scale parameters are exactly like in the GtkRadiant surface properties. The Flags field only uses one bit to describe whether this is a detail brush or not.

The points ABC are not necessarily actual brush vertices. They CAN be, but often they aren't. To get the actual vertices, you need to intersect the planes with others.

Thanks for the info. I intend to write a small app that turns hight maps into tri-soups (or maybe quad-soups). This makes it a bit easier for me.

[tZork]

gensurf can do that, its included as a plugin in gtkradiant 1.4 and 1.5. http://tarot.telefragged.com/gensurf/

[divVerent]

Where is it in 1.5? I found a directory gtkgensurf that is not part of the default build, and when trying to add it, it missed a header file qetypes.h that was nowhere in GtkRadiant. Can you tell me how to enable GtkGenSurf in GtkRadiant 1.5?

[tZork]

Where is it in 1.5? I found a directory gtkgensurf that is not part of the default build, and when trying to add it, it missed a header file qetypes.h that was nowhere in GtkRadiant. Can you tell me how to enable GtkGenSurf in GtkRadiant 1.5?

Hmm yes.. my current build lacks it too. i know i had it in 1.5 at some point i have no idea how to get it working again. The standalone one should work, but it looks like the only d/l avaiable on the site it a windows executable :/ Seems i jumped the gun here, sorry.

[Rad Ished]

My macradiant 1.4 build has it

[divVerent]

Darn. So this is AGAIN something that is in 1.4 only. Too bad there is no 1.4 for Linux 64bit. Hopefully soon.

[tZork]

I seriosly hope the source of 1.4 will be avaiable soon. imo its superior to 1.5 in most ways. at least for my use. beeing able to tinker with the souce and build it for *nix would be very nice indeed.

[KadaverJack]

I seriosly hope the source of 1.4 will be avaiable soon. imo its superior to 1.5 in most ways. at least for my use. beeing able to tinker with the souce and build it for *nix would be very nice indeed.

it already is, but i have no idea whether it's in a usable state...
http://www.qeradiant.com/cgi-bin/trac.c ... eroRadiant